Predicting Stock Price Using The Quality of Narrative Disclosure About Risks (an Experimental Study 0n The Markets of The Gulf Cooperation Council Countries)
Keywords:
Risk narrative disclosure quality, stock price prediction, GCC stock markets, artificial neural networks.Abstract
The study aims to examine the predictive power of risk narrative disclosure quality in explaining stock prices among non‑financial firms listed in the stock markets of the Gulf Cooperation Council (GCC) countries, using artificial intelligence techniques—specifically, an Artificial Neural Network (ANN) model. The quality of narrative disclosure was measured through a content‑analysis approach based on a multidimensional index comprising four main components: coverage, quantity, richness, and time orientation, in order to construct a comprehensive indicator that reflects the level of risk narrative disclosure in annual reports.
The study relies on panel data for 421 non‑financial firms listed in GCC stock markets over the period 2010–2024. A Multilayer Perceptron (MLP) neural network model was employed to develop a predictive model for stock prices using the quality of risk narrative disclosure as an input variable.
The neural network results indicate that the quality of risk narrative disclosure contributes positively to predicting stock prices; however, the net effect value was only 0.004, which is relatively low and suggests that the variable’s influence is weak and insufficient to serve as a primary determinant in stock‑price prediction. Accordingly, while narrative disclosure enhances the information environment and reduces information asymmetry—thereby enabling investors to form more accurate expectations regarding future risks—it remains a complementary rather than a central informational source in the pricing process within GCC stock markets.